Determining the basic psychometric properties of the Greek KDQOL-SF

The aim of this study was to determine the basic psychometric properties, i.e. reliability and validity, of the Greek version of the Kidney Disease Quality of Life Short Form (KDQOL-SF). The instrument was self-administered to a homogenous group of 665 end stage renal disease patients in 20 dialysis units throughout Greece and the overall response rate was 72.6%. Reliability was demonstrated by Cronbach's alpha exceeding the recommended minimum value of 0.70 in all, except one, scales. Tests of item-internal consistency, after correction for overlap, resulted in correlations between items and their hypothesized scales, which exceeded the 0.40 standard in 94.5% of the cases. Item discriminant validity tests indicated 100% scaling success for six out of eight generic and disease-targeted scales. Validity was supported by the confirmation of expected correlations between scales and the overall health-rating item included in the instrument and with sociodemographic and self-reported health variables. Multiple stepwise linear regression analysis demonstrated that all disease-targeted scales were important predictors of SF-36 general health scales and the variance explained ranged from 37% to 57%. Overall, the psychometric properties of the KDQOL-SF, resulting from this first-time administration of the instrument to a Greek dialysis population, were good and the disease targeted scales were informative and of high internal consistency reliability. Cross-sectional construct validity is demonstrated, despite the lack of external validity criteria based on clinical ratings of severity. The results support administering the Greek KDQOL-SF in studies evaluating dialysis therapy and contribute to transnational comparison of findings.